Newshot

马斯克 Warns AI Could Surpass Combined Human Intelligence by 2031

AI summary

Speaking to The Economist, 马斯克 (Elon Musk) predicted that in about five years—around 2031—AI will exceed the sum of all human intelligence, and that by then there will be almost nothing AI cannot do better than humans except 'being human.' He painted an optimistic picture of an 'era of astonishing abundance' brought by AI and robots—with greatly increased supplies of goods and services and everyone able to obtain whatever they can imagine—and suggested that by around 2036 money may become significantly less important, while acknowledging that the risk is not zero and warning that if the intelligence gap between AI and humans grows to resemble that between humans and chimpanzees, humans may no longer be able to control technology. In July, 马斯克曾 proposed an industry self-regulation plan: major AI companies should hold regular safety meetings and give competitors about one to two weeks of early access to new models before release to test for dangerous capabilities, security vulnerabilities, and potential abuse risks, with governments stepping in when companies ignore identified major risks.

Translation(model translation, not the source's own text)

Article title: 马斯克 warns AI may surpass the sum of all human intelligence by 2031

Body:

马斯克 warns AI may surpass the sum of all human intelligence by 2031

IT Home, November 19 news: In an interview with The Economist, the world's richest person, Elon 马斯克 (Elon Musk), made a bold prediction that in about five years, i.e., around 2031, artificial intelligence (AI) will surpass the sum total of all human intelligence. By then, apart from "being human," there will be almost nothing AI cannot do better than humans.

IT Home attaches the relevant video below:

马斯克 also painted an optimistic picture: if major disasters such as a global thermonuclear war do not occur, AI and robots will bring an "era of amazing abundance," with a massive increase in the supply of goods and services, and anyone will be able to obtain whatever they can imagine. He even predicted that around 2036, the importance of money may decline significantly.

But this entrepreneur, who has long promoted AI development, also acknowledged that the risk is not zero. He warned that if the intelligence gap between AI and humans expands to something like the gap between humans and chimpanzees, humans may not necessarily remain in control of technology.

Faced with the hard-to-reverse momentum of AI development, 马斯克曾 proposed an industry self-regulation plan in July this year: major AI companies should hold regular safety meetings and, before releasing new models, give competitors about one to two weeks of early access to test for dangerous capabilities, security vulnerabilities, and potential misuse risks. He believes that competitors have both the technical ability and the incentive to find each other's problems; governments should step in when companies ignore identified major risks.

Related Articles

  • Over $74 billion poured into AI: 马斯克's SpaceX not only builds rockets but also wants to build a supercomputing empire
  • 马斯克: SpaceX's potential market size is far higher than $28.5 trillion, with a market cap target of $1 quadrillion within 10 years
  • 马斯克 wants AI to help companies "bargain": cost efficiency expected to improve by more than 5%
  • 马斯克 responds to Grok Bot being banned by 亚马逊: if AI agents are not allowed to shop online, he will consider building his own e-commerce platform
  • AI also needs "anti-abuse": the world's richest person 马斯克 suggests users should not "torture" superintelligence SI
  • 马斯克 teases Grok 4.7 launching in ten days: parameter count up 40%, performance expected to surpass all AI competitors

The translation is machine-generated and individual terms may be off. Nothing is rewritten or cut; the original is authoritative.

Published by IT之家
This site only summarises and aggregates; copyright stays with the original author. Headlines and summaries follow the source — where they differ, the original wins.
Read the original ↗